The biological activities of thyroid hormone (T3) are mediated by thyroid hormone receptors (TRs). There are two TR genes, alpha and beta, which yield T3-binding TR alpha 1, TR beta 1, TR beta 2 and TR beta 3 isoforms by alternative splicing of the primary transcripts. In the past years, we have been using the powerful tool of mouse genetics not only to understand the molecular mechanisms of thyroid hormone action in vivo, but also to delineate the molecular basis of diseases due to mutations of TRs. Creating a mouse model of resistance to thyroid hormone (RTH) RTH is a disease due to mutations of the TR beta gene. A mutation derived from an RTH patient at NIH (TR beta PV) was targeted to the TR beta gene locus using homologous recombination and the Cre/lox system (TR beta PV mouse). TR beta PV mice faithfully reproduce the human RTH. Using this mouse model, it has become possible to study several critical, clinically relevant issues that previously could not be studied. We showed that RTH symptoms are caused by the interference of mutant TR beta with the transcriptional activity of wild-type TR alpha 1 and TR beta in vivo. We also discovered that variable phenotypic expression in RTH patients is dictated by tissue-dependent abundance of TR beta and TR alpha 1 isoforms, the promoter context of T3 target genes and tissue-dependent expression of coactivators, such as the steroid hormone receptor coactivator-1 (SRC-1). In addition, we found that the mild phenotype frequently observed in heterozygous RTH patients is due to the compensatory effect of wild-type TR alpha 1. Identifying TR betaPV/PV mice as a model of thyroid carcinogenesis An unexpected but remarkable discovery is that TR betaPV/PV mice spontaneously develop thyroid carcinoma with sequential capsular invasion, vascular invasion, anaplasia, and eventually metastasis. The lack of valid mouse models with metastatic spread has impeded the progress in understanding of the molecular events of thyroid cancer invasion and metastasis. Thus, this mouse model provides an unprecedented opportunity to study gene alterations during thyroid carcinogenesis. Using cDNA microarrays, we found altered expression of 20 named genes involved in tumor induction and progression, 16 named genes in invasion and metastasis, and 3 genes in cell proliferation and cell cycle regulation. We have also identified several altered signaling pathways that contribute to thyroid carcinogenesis of TR betaPV/PV mice. We discovered that the peroxisome proliferator activated receptor could be tested as a potential molecular target for prevention and treatment of follicular thyroid carcinoma. Discovering TR betaPV/PVmice as a model of pituitary tumorigenesis Thyrotropin (TSH)-secreting tumors (TSHomas) are pituitary tumors that constitutively secrete TSH. More than 300 cases have been reported. TSHomas are usually large at diagnosis and are associated with headaches, visual field disturbances, and deficiency in other pituitary hormones. Because diagnosis occurs late in the natural course, curative surgical resection of TSHomas remains under 50%. The molecular genetics underlying TSHomas are not well understood. We discovered that TR betaPV/PVmice spontaneously develop TSHomas; thus, providing the first mouse model to elucidate molecular genetic events underlying tumorigenesis of the pituitary and to identify potential molecular targets for treatment. Creating a novel mouse model (TR alpha PV mice) to elucidate the genetic basis of human dwarfism The intriguing observation that no mutations of the TR alpha gene have ever been identified in RTH patients has perplexed investigators for years and raised the fundamental issue of whether mutations of the TR alpha gene are lethal or can cause other human diseases. We have successfully created the first TR alpha knockin mutant mouse, which unequivocally resolved this fundamental issue. The mutation of one allele of the TR alpha gene is not lethal, but results in dwarfism and other abnormalities that are clearly distinct from RTH. The demonstration that the mutation of the TR alpha gene leads to dwarfism has generated significant excitement in the field. This mouse model is being used to study possible TR alpha mutations in children with growth abnormalities.